U.S. flag

An official website of the United States government, Department of Justice.

Post-Blast Explosives Attribution

NCJ Number
Joshua Dettman, Ph.D.
Date Published
43 pages

This is the Final Report on the findings and design of a research project with the primary goal of determining whether relevant conserved signatures of an explosive source can be recovered post-blast and matched to pre-blast signatures.



Upon examination of project results, the research team concluded that the overall results of the research show promise for the process used to detect and identify signatures for attribution in post-blast residues. The explosive type that yielded the most samples with useful signature data was AN-AL. Trace element data were collected for nearly all samples measured, and all 30 samples sent for analyses with infrared mass spectrometry (IRMS) were at detectable levels for both isotopes measured (oxygen and nitrogen). Although the number of samples alone does not provide conclusive information, the greater number of data points does provide a stronger dataset and greater confidence in the result. For the organic explosives tested, post-blast RDX samples did not yield any detectable amounts of explosive; therefore, no signature data were obtained sufficient to draw any conclusions. Given that RDX is a high-order explosive and nearly all recoverable explosive material is consumed, it cannot be conclusively determined from this test whether post-blast attribution is possible with RDX. TNT, despite being a high-order explosive like RDX, did yield 26 post-blast samples with detectable levels of explosive. HPLC-MS profiling identified four compounds that were present in both the pre-blast and all 26 post-blast samples. They were not present in any blanks or negative controls. A key limitation encountered during the study was obtaining recoverable amounts of the high-order explosive RDX and TNT. The rapid consumption of explosive material resulted in no recovered samples of RDX and only 26 out of 108 of TNT. Other potential factors for poorly recovered material included the place positions of sample collection devices on the sample grid and the number of total sample collection devices. 20 figures, 6 tables, and 26 references


Date Published: January 1, 2022